CN104933204A - Intelligent network response method and device - Google Patents
Intelligent network response method and device Download PDFInfo
- Publication number
- CN104933204A CN104933204A CN201510416844.9A CN201510416844A CN104933204A CN 104933204 A CN104933204 A CN 104933204A CN 201510416844 A CN201510416844 A CN 201510416844A CN 104933204 A CN104933204 A CN 104933204A
- Authority
- CN
- China
- Prior art keywords
- language material
- original language
- user
- intelligent network
- condition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3325—Reformulation based on results of preceding query
- G06F16/3326—Reformulation based on results of preceding query using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
Abstract
The invention relates to an intelligent network response method and device. Original linguistic data are stored according to the preset condition and the preset class; after consult problems proposed by a user are received, the original linguistic data which correspond to the class and meet the condition are selected from the stored original linguistic data; commodity information consulted in the consult problems is obtained, and reserved positions in the selected original linguistic data are then filled with the commodity information so as to form answer texts; the answer texts are output, then modifications on the answer texts are received, and the modified answer texts are transmitted to the user. According to the intelligent network response method and device, a knowledge graph is formed, the linguistic data are selected according to the class and the condition corresponding to the consult problems proposed by the user, the mode that at present, a common problem and the problem class are simply set, and then fixed words are used for answering is broken through, classification is achieved, answering is made according to the individuality of the user, more consult problems can be automatically answered, and the higher answering accuracy is achieved.
Description
Technical field
The present invention relates to field of computer technology, the method and apparatus of particularly a kind of intelligent network response.
Background technology
Customer service system is that auxiliary enterprises contact staff receives client, answer process the computer system of client's counseling problem, this customer service system only has the function that user and online customer service carry out seeking advice from, if customer service is not online or festivals or holidays, the online service of user is just interrupted; At present along with customer volume is increasing, within 24 hours, Online reference service module is also essential; Intelligent response system is well positioned to meet the demand of enterprise for this reason; To the service of industry promote one leap.
But the weak point that existing customer service system exists cannot carry out personalization response for the feature of user.For the user of different situations (such as sex, age, occupation etc.), the answer provided is identical.Such one is be difficult to obtain better Consumer's Experience, and two is carry out, in the linguistic context promoted, have impact on distribution effect at needs.Response mode based on problem category can only belong to the problem of classification corresponding to this answer by artificial predefined answer answer, and artificial predefined answer expression way is limited and relatively fixing, and the accuracy of answer is limited maybe cannot answer.Scheme based on the retrieval in question and answer storehouse is subject to the restriction of question and answer storehouse scale, cannot answer or provide inaccurate answer for the counseling problem exceeding question and answer storehouse.
Summary of the invention
For above problem, the invention provides the method and apparatus of a kind of intelligent network response, contribute to more counseling problem is answered automatically, and have higher response precision.
For achieving the above object, according to an aspect of the present invention, the method for a kind of intelligent network response is provided.
The method of intelligent network response of the present invention comprises:
A, preserve original language material according to pre-conditioned and pre-set categories; Describedly pre-conditionedly comprise the condition pre-set according to merchandise news, described pre-set categories corresponding to the classification of the problem pre-set, and comprises the reserved position for filling merchandise news in described original language material;
B, after receiving the counseling problem that user proposes, determine classification belonging to this counseling problem and the condition that meets, choose corresponding to this classification from the original language material of preservation and meet the original language material of this condition;
C, obtain the merchandise news that described counseling problem seeks advice from and be then filled into the reserved location in the original language material chosen thus form answer text.
D, export described answer text, then accept the amendment to described answer text, then send amended described answer text to described user.
Alternatively, before described steps A, also comprise:
A1, obtain multinomial original language material; By described pre-set categories, cluster is carried out to described multinomial original language material and preserves; For each the original language material belonged to after cluster in the original language material of a class, pre-conditionedly preserve this original language material respectively by described.
A2, a reception original language material, then preserve this original language material according to the pre-conditioned and pre-set categories of specifying.
Alternatively, describedly pre-conditionedly the condition pre-set according to the customized information of user is also comprised; The reserved position for filling customized information is also comprised in described original language material;
Alternatively, also comprise question sentence language material in described original language material, this question sentence language material is for obtaining the customized information of user.
Alternatively, also comprise: classification process is carried out with the conventional Category of consulting obtaining this designated user, then using the customized information of this conventional Category of consulting as this designated user by the Category of consulting preset to the history counseling problem of designated user.
Alternatively, the textual entry of the consulting conversation recording of designated user is added up, if comprise the ratio of specifying the textual entry of vocabulary to reach the textual entry sum of the consulting conversation recording of this designated user to reach preset value, then the customized information of customized information as this designated user of described appointment vocabulary will be corresponded to.
According to a further aspect in the invention, the device of a kind of intelligent network response is provided.
The device of intelligent network response of the present invention comprises:
Preserve module, for preserving original language material according to pre-conditioned and pre-set categories; Describedly pre-conditionedly comprise the condition pre-set according to merchandise news, described pre-set categories corresponding to the classification of the problem pre-set, and comprises the reserved position for filling merchandise news in described original language material;
Language material chooses module, for after receiving the counseling problem that user proposes, determine classification belonging to this counseling problem and the condition that meets, choose corresponding to this classification from the original language material of preservation and meet the original language material of this condition;
Merchandise news acquisition module, for obtaining the merchandise news that described counseling problem is seeked advice from;
Packing module, forms answer text for the reserved location that described merchandise news is filled in original language material that described language material chooses selected by module.
Output module, described output module, for exporting described answer text, then accepts the amendment to described answer text; And then send amended described answer text to described user.
Alternatively, also comprise knowledge mapping and build module, for:
M1, obtain multinomial original language material; By described pre-set categories, cluster is carried out to described multinomial original language material and preserves; For each the original language material belonged to after cluster in the original language material of a class, pre-conditionedly preserve this original language material respectively by described.
M2, a reception original language material, then preserve this original language material according to the pre-conditioned and pre-set categories of specifying.
M3, the textual entry of the consulting conversation recording of designated user to be added up, if comprise the ratio of specifying the textual entry of vocabulary to reach the textual entry sum of the consulting conversation recording of this designated user to reach preset value, then the customized information of customized information as this designated user of described appointment vocabulary will be corresponded to.
M4, classification process is carried out with the conventional Category of consulting obtaining this designated user, then using the customized information of this conventional Category of consulting as this designated user by the Category of consulting preset to the history counseling problem of designated user.
Alternatively, describedly pre-conditionedly the condition pre-set according to the customized information of user is also comprised; The reserved position for filling customized information is also comprised in described original language material.
The invention has the beneficial effects as follows:
Original language material is preserved according to pre-conditioned and pre-set categories, thus constitute knowledge mapping, the classification corresponding according to the counseling problem of user's proposition and condition choose language material, and the reserved location of filling in language material is to form answer text, this mode breaches the existing mode simply arranging common question and Question Classification and then carry out according to fixing term answering, achieve classification and answer according to user personality, more counseling problem is answered automatically, and has higher response precision.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of the method for replying according to the intelligent network of the embodiment of the present invention;
Fig. 2 is the schematic diagram of the main modular of the device of replying according to the intelligent network of the embodiment of the present invention.
Embodiment
Below in conjunction with the drawings and specific embodiments, the invention will be further described.
Embodiment 1
As shown in Figure 1, a kind of method of intelligent network response comprises:
A, preserve original language material according to pre-conditioned and pre-set categories; Describedly pre-conditionedly comprise the condition pre-set according to merchandise news, described pre-set categories corresponding to the classification of the problem pre-set, and comprises the reserved position for filling merchandise news in described original language material;
B, after receiving the counseling problem that user proposes, determine classification belonging to this counseling problem and the condition that meets, choose corresponding to this classification from the original language material of preservation and meet the original language material of this condition;
C, obtain the merchandise news that described counseling problem seeks advice from and be then filled into the reserved location in the original language material chosen thus form answer text.
D, export described answer text, then accept the amendment to described answer text, then send amended described answer text to described user.
Before described steps A, also comprise:
A1, obtain multinomial original language material; By described pre-set categories, cluster is carried out to described multinomial original language material and preserves; For each the original language material belonged to after cluster in the original language material of a class, pre-conditionedly preserve this original language material respectively by described.
A2, a reception original language material, then preserve this original language material according to the pre-conditioned and pre-set categories of specifying.
The answer of some common question can be set in businessman's customer service system, as " sending out what express delivery ", " bag postal ", " what reimbursement address is ", also can safeguard the merchandise newss such as the size specification of clothing commodity.These answers have clear and definite classification information, such as above-mentioned three problems belong to " logistics class " information, and the text of above-mentioned three problems can as common question, the particular problem that client proposes can identify this particular problem by the mode such as word segmentation result and text matches and correspond to which common question, such as particular problem " may I ask reimbursement address is what ", " what reimbursement address is; could you tell me ", " which reimbursement address is " etc., as many asly asking common question " what reimbursement address is ".
Due to the common question limited amount that can arrange, businessman also can arrange unlimited self-defined question and answer.Such as problem is " may I ask ", corresponding answer be ", may I ask and need anything to help " or " hello " etc.These question and answer do not have classification information, only comprise corresponding problem and answer.
When client modules receives the counseling problem of user's transmission, this counseling problem is passed into system, and system imports this counseling problem into JIMI after filtering inefficiency according to keyword.JIMI can use the machine learning algorithm such as natural language processing technique and support vector machine, logistic regression, intention assessment such as Chinese word segmentation, word sense disambiguation, semantic error correction, Entity recognition, intention assessment by Question Classification, then judge, if customer issue corresponds to certain common question in certain problem category of businessman's setting, then JIMI can provide the answer of this common question.Such as, when user asks " which reimbursement address is ", first JIMI determines that the classification belonging to this problem is " logistics class ", then determine that the common question corresponding to this problem is " what reimbursement address is ", then the answer (such as: " reimbursement address is Shengli road No. 22 1st floor 1 Room 402 ") corresponding to this common question pre-set is returned to user.If the above-mentioned judgement of JIMI finds that customer issue corresponds to certain common question in certain problem category of businessman's setting, then JIMI retrieves the self-defined problem of businessman; As retrieved not yet, then JIMI provides the personalized answers in " thousand people thousand faces " according to the personalized descriptor of user.
Such as, the personalized descriptor of party A-subscriber and party B-subscriber is respectively:
Party A-subscriber: female, 22 years old, university student, unmarried, height 165cm, responsive to sales promotion, Silver Subscriber;
Party B-subscriber: man, 28 years old, white collar, married, there is child, 181cm, high to evaluation attention rate, diamond user.
The problem of party A-subscriber and party B-subscriber's consulting is all that such as one of them counseling problem is about apple iPhone6A158916G version 4G cell phone: " camera how many pixels ".Now JIMI to the answer of party A-subscriber is: " camera is 8,000,000 pixels, and secondary camera is 1,200,000 pixels.Taking advantage of young to oneself taking a little photo more, keeping eternal youth and beauty here ".And JIMI to the answer of party B-subscriber is: " camera is 8,000,000 pixels, and secondary camera is 1,200,000 pixels.Pick up it at any time and clap a photo to your child, keep each splendid moment here ".And for example wherein another counseling problem is: " screen is much ".Now JIMI to the answer of party A-subscriber is: " screen 4.7 cun, resolution 1334 × 750.Brand-new Retina high-definition display screen, bed of lying is seen what the most refreshing of TV play ".And JIMI to the answer of party B-subscriber is: " screen 4.7 cun, resolution 1334 × 750.Based on your height of 1 meter 81, the iPhone6plus of 5.5 cun of screens may be more suitable for you "; Although can find out it is the mobile phone of seeking advice from same model, the answer of JIMI to different user is not quite similar, and forms personalized answers.
Can arrange some question sentence language materials in original language material, these question sentence language materials are in a manner mentioned above equally, and category and condition are preserved, and such JIMI can utilize these question sentence language materials to obtain the customized information of user.Such as following dialogue:
User: I wears much?
JIMI: may I ask your height and weight how many?
User: height 170, body weight 60
JIMI: this part clothes has M code, and you wear just in time suitable.
Utilize above-mentioned dialogue, the user personalized information that JIMI achieves based on Entity recognition obtains.In general, Entity recognition can adopt all texts in grammar tree analysis session, and each tree node all judges whether to satisfy condition with regular expression, and general effect is equivalent to the regular expression combining multiple complexity.
The customized information of user can also comprise user and commonly use Category of consulting.Determining that user commonly uses Category of consulting and is, classification process can carried out to the history counseling problem of designated user by the Category of consulting preset, separately can be aided with certain rule and filter, thus obtain the conventional Category of consulting of this designated user.This mode is the customized information obtaining user based on sorting algorithm.The customized information of user separately can be obtained based on text matches and statistics.Can be specifically that the textual entry of the consulting conversation recording of designated user is added up, if comprise the ratio of specifying the textual entry of vocabulary to reach the textual entry sum of the consulting conversation recording of this designated user to reach preset value, then the customized information of customized information as this designated user of above-mentioned appointment vocabulary will be corresponded to.Such as counting user has the statement of more than 50% to include default uncivil term, then can determine this user's irritability.Be somebody's turn to do the customized information that " irritability " is this user.
Embodiment 2
As shown in Figure 2, provide the device of a kind of intelligent network response, this device comprises:
Preserve module, for preserving original language material according to pre-conditioned and pre-set categories; Describedly pre-conditionedly comprise the condition pre-set according to merchandise news, described pre-set categories corresponding to the classification of the problem pre-set, and comprises the reserved position for filling merchandise news in described original language material;
Language material chooses module, for after receiving the counseling problem that user proposes, determine classification belonging to this counseling problem and the condition that meets, choose corresponding to this classification from the original language material of preservation and meet the original language material of this condition;
Merchandise news acquisition module, for obtaining the merchandise news that described counseling problem is seeked advice from;
Packing module, forms answer text for the reserved location that described merchandise news is filled in original language material that described language material chooses selected by module.
Output module, described output module, for exporting described answer text, then accepts the amendment to described answer text; And then send amended described answer text to described user.
According to the technical scheme of the embodiment of the present invention, original language material is preserved according to pre-conditioned and pre-set categories, thus constitute knowledge mapping, the classification corresponding according to the counseling problem of user's proposition and condition choose language material, and the reserved location of filling in language material is to form answer text, this mode breaches the existing mode simply arranging common question and Question Classification and then carry out according to fixing term answering, achieve classification and answer according to user personality, more counseling problem is answered automatically, and has higher response precision; And then the instantaneous value that can obtain merchandise news when filling the reserved location in language material from the database of e-commerce platform fills, thus can reply with latest data user.Have employed multiple means when realizing knowledge mapping, such as, obtain multinomial original language material and then classify, a large amount of original language material can be obtained efficiently; Also the language material of those inconvenience by computing machine automatic acquisition can be set specially by developer; In acquisition user personalized information, have employed based on Entity recognition, based on sorting algorithm, based on means such as text matches level statistics, these all contribute to obtaining complete same user personalized information.
Below describe ultimate principle of the present invention in conjunction with specific embodiments, in apparatus and method of the present invention, obviously, each parts or each step can decompose and/or reconfigure.These decompose and/or reconfigure and should be considered as equivalents of the present invention.Further, the step performing above-mentioned series of processes can order naturally following the instructions perform in chronological order, but does not need necessarily to perform according to time sequencing.Some step can walk abreast or perform independently of one another.
Above-mentioned embodiment, does not form limiting the scope of the invention.It is to be understood that depend on designing requirement and other factors, various amendment, combination, sub-portfolio can be there is and substitute in those skilled in the art.Any amendment done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within scope.
Claims (8)
1. a method for intelligent network response, the method comprises:
A, preserve original language material according to pre-conditioned and pre-set categories; Describedly pre-conditionedly comprise the condition pre-set according to merchandise news, described pre-set categories corresponding to the classification of the problem pre-set, and comprises the reserved position for filling merchandise news in described original language material;
B, after receiving the counseling problem that user proposes, determine classification belonging to this counseling problem and the condition that meets, choose corresponding to this classification from the original language material of preservation and meet the original language material of this condition;
C, obtain the merchandise news that described counseling problem seeks advice from and be then filled into the reserved location in the original language material chosen thus form answer text.
D, export described answer text, then accept the amendment to described answer text, then send amended described answer text to described user.
2. the method for intelligent network response according to claim 1, is characterized in that: before described steps A, also comprise:
A1, obtain multinomial original language material; By described pre-set categories, cluster is carried out to described multinomial original language material and preserves; For each the original language material belonged to after cluster in the original language material of a class, pre-conditionedly preserve this original language material respectively by described.
A2, a reception original language material, then preserve this original language material according to the pre-conditioned and pre-set categories of specifying.
3. the method for intelligent network according to claim 1 response, is characterized in that: describedly pre-conditionedly also comprise the condition pre-set according to the customized information of user; The reserved position for filling customized information is also comprised in described original language material.
4. the method for intelligent network response according to claim 1, it is characterized in that: also comprise question sentence language material in described original language material, this question sentence language material is for obtaining the customized information of user.
5. the method for intelligent network response according to claim 1, it is characterized in that: also comprise: classification process is carried out with the conventional Category of consulting obtaining this designated user, then using the customized information of this conventional Category of consulting as this designated user by the Category of consulting preset to the history counseling problem of designated user.
6. a device for intelligent network response, comprising:
Preserve module, for preserving original language material according to pre-conditioned and pre-set categories; Describedly pre-conditionedly comprise the condition pre-set according to merchandise news, described pre-set categories corresponding to the classification of the problem pre-set, and comprises the reserved position for filling merchandise news in described original language material;
Language material chooses module, for after receiving the counseling problem that user proposes, determine classification belonging to this counseling problem and the condition that meets, choose corresponding to this classification from the original language material of preservation and meet the original language material of this condition;
Merchandise news acquisition module, for obtaining the merchandise news that described counseling problem is seeked advice from;
Packing module, forms answer text for the reserved location that described merchandise news is filled in original language material that described language material chooses selected by module.
Output module, described output module, for exporting described answer text, then accepts the amendment to described answer text; And then send amended described answer text to described user.
7. the device of intelligent network response according to claim 6, is characterized in that: also comprise knowledge mapping and build module, for:
M1, obtain multinomial original language material; By described pre-set categories, cluster is carried out to described multinomial original language material and preserves; For each the original language material belonged to after cluster in the original language material of a class, pre-conditionedly preserve this original language material respectively by described.
M2, a reception original language material, then preserve this original language material according to the pre-conditioned and pre-set categories of specifying.
M3, the textual entry of the consulting conversation recording of designated user to be added up, if comprise the ratio of specifying the textual entry of vocabulary to reach the textual entry sum of the consulting conversation recording of this designated user to reach preset value, then the customized information of customized information as this designated user of described appointment vocabulary will be corresponded to.
M4, classification process is carried out with the conventional Category of consulting obtaining this designated user, then using the customized information of this conventional Category of consulting as this designated user by the Category of consulting preset to the history counseling problem of designated user.
8. the device of intelligent network according to claim 6 response, is characterized in that: describedly pre-conditionedly also comprise the condition pre-set according to the customized information of user; The reserved position for filling customized information is also comprised in described original language material.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510416844.9A CN104933204A (en) | 2015-07-15 | 2015-07-15 | Intelligent network response method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510416844.9A CN104933204A (en) | 2015-07-15 | 2015-07-15 | Intelligent network response method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104933204A true CN104933204A (en) | 2015-09-23 |
Family
ID=54120371
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510416844.9A Pending CN104933204A (en) | 2015-07-15 | 2015-07-15 | Intelligent network response method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104933204A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105450781A (en) * | 2016-01-12 | 2016-03-30 | 李劲杰 | Interactive question and answer method, platform, client and server |
| CN107016044A (en) * | 2017-02-17 | 2017-08-04 | 阿里巴巴集团控股有限公司 | A kind of method and device of data visualization processing |
| CN107784051A (en) * | 2016-12-15 | 2018-03-09 | 平安科技（深圳）有限公司 | Online customer service answering system and method |
| CN107807960A (en) * | 2017-09-30 | 2018-03-16 | 平安科技（深圳）有限公司 | Intelligent customer service method, electronic installation and computer-readable recording medium |
| CN108428148A (en) * | 2018-01-29 | 2018-08-21 | 厦门快商通信息技术有限公司 | Active smart phone marketing method and system |
| CN109214854A (en) * | 2018-07-10 | 2019-01-15 | 淘度客电子商务股份有限公司 | A kind of self-help service for user method applied to electric business platform |
| CN109829041A (en) * | 2018-12-25 | 2019-05-31 | 出门问问信息科技有限公司 | Question processing method and device, computer equipment and computer readable storage medium |
| CN111178922A (en) * | 2018-11-09 | 2020-05-19 | 阿里巴巴集团控股有限公司 | Service providing method, virtual customer service generating method, device and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130073277A1 (en) * | 2011-09-21 | 2013-03-21 | Pket Llc | Methods and systems for compiling communication fragments and creating effective communication |
| CN103177039A (en) * | 2011-12-26 | 2013-06-26 | 中国移动通信集团北京有限公司 | Data processing method and data processing device |
| CN103559619A (en) * | 2013-11-12 | 2014-02-05 | 北京京东尚科信息技术有限公司 | Response method and system for garment size information |
| CN104731895A (en) * | 2015-03-18 | 2015-06-24 | 北京京东尚科信息技术有限公司 | Auto-answer method and device |
-
2015
- 2015-07-15 CN CN201510416844.9A patent/CN104933204A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130073277A1 (en) * | 2011-09-21 | 2013-03-21 | Pket Llc | Methods and systems for compiling communication fragments and creating effective communication |
| CN103177039A (en) * | 2011-12-26 | 2013-06-26 | 中国移动通信集团北京有限公司 | Data processing method and data processing device |
| CN103559619A (en) * | 2013-11-12 | 2014-02-05 | 北京京东尚科信息技术有限公司 | Response method and system for garment size information |
| CN104731895A (en) * | 2015-03-18 | 2015-06-24 | 北京京东尚科信息技术有限公司 | Auto-answer method and device |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105450781A (en) * | 2016-01-12 | 2016-03-30 | 李劲杰 | Interactive question and answer method, platform, client and server |
| CN107784051A (en) * | 2016-12-15 | 2018-03-09 | 平安科技（深圳）有限公司 | Online customer service answering system and method |
| CN107784051B (en) * | 2016-12-15 | 2020-09-18 | 平安科技（深圳）有限公司 | Online customer service response system and method |
| CN107016044A (en) * | 2017-02-17 | 2017-08-04 | 阿里巴巴集团控股有限公司 | A kind of method and device of data visualization processing |
| CN107016044B (en) * | 2017-02-17 | 2020-07-24 | 阿里巴巴集团控股有限公司 | Data visualization processing method and device |
| CN107807960A (en) * | 2017-09-30 | 2018-03-16 | 平安科技（深圳）有限公司 | Intelligent customer service method, electronic installation and computer-readable recording medium |
| WO2019062078A1 (en) * | 2017-09-30 | 2019-04-04 | 平安科技（深圳）有限公司 | Smart customer service method, electronic apparatus and computer-readable storage medium |
| CN108428148A (en) * | 2018-01-29 | 2018-08-21 | 厦门快商通信息技术有限公司 | Active smart phone marketing method and system |
| CN109214854A (en) * | 2018-07-10 | 2019-01-15 | 淘度客电子商务股份有限公司 | A kind of self-help service for user method applied to electric business platform |
| CN111178922A (en) * | 2018-11-09 | 2020-05-19 | 阿里巴巴集团控股有限公司 | Service providing method, virtual customer service generating method, device and electronic equipment |
| CN109829041A (en) * | 2018-12-25 | 2019-05-31 | 出门问问信息科技有限公司 | Question processing method and device, computer equipment and computer readable storage medium |
| CN109829041B (en) * | 2018-12-25 | 2021-06-29 | 出门问问信息科技有限公司 | Question processing method and device, computer equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104731895A (en) | Auto-answer method and device | |
| CN104933204A (en) | Intelligent network response method and device | |
| US11599714B2 (en) | Methods and systems for modeling complex taxonomies with natural language understanding | |
| US20240078386A1 (en) | Methods and systems for language-agnostic machine learning in natural language processing using feature extraction | |
| US10992609B2 (en) | Text-messaging based concierge services | |
| CN109299994B (en) | Recommendation method, device, equipment and readable storage medium | |
| WO2019037391A1 (en) | Method and apparatus for predicting customer purchase intention, and electronic device and medium | |
| CN111260437B (en) | Product recommendation method based on commodity-aspect-level emotion mining and fuzzy decision | |
| CN111784455A (en) | Article recommendation method and recommendation equipment | |
| US10127619B2 (en) | Determination of targeted food recommendation | |
| CN107291840B (en) | User attribute prediction model construction method and device | |
| CN107656948A (en) | The problem of in automatically request-answering system clustering processing method and device | |
| CN106997347A (en) | Information recommendation method and server | |
| CN110263161B (en) | Information processing method, device and equipment | |
| CN107025283A (en) | The answer method and system of candidate answers sequence are carried out based on subscriber data | |
| CN112733042A (en) | Recommendation information generation method, related device and computer program product | |
| CN113886691A (en) | Intelligent recommendation method and device based on historical data, electronic equipment and medium | |
| CN111353838A (en) | Method and device for automatically checking commodity category | |
| CN113407677A (en) | Method, apparatus, device and storage medium for evaluating quality of consultation session | |
| CN110852785B (en) | User grading method, device and computer readable storage medium | |
| CN107357782A (en) | One kind identification user's property method for distinguishing and terminal | |
| US20230015090A1 (en) | Systems and Methods for Dynamically Classifying Products and Assessing Applicability of Product Regulations | |
| CN112766779B (en) | Information processing method, computer device, and storage medium | |
| Moody et al. | Towards explainable summary of crowdsourced reviews through text mining | |
| CN115408611A (en) | Menu recommendation method and device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150923 |